SIAMESE VERIFICATION FRAMEWORK FOR AUTISM IDENTIFICATION DURING INFANCY USING CORTICAL PATH SIGNATURE FEATURES.

SIAMESE VERIFICATION FRAMEWORK FOR AUTISM IDENTIFICATION DURING INFANCY USING CORTICAL PATH SIGNATURE FEATURES.
复制标题

DOI:
10.1109/isbi45749.2020.9098385
复制
发表时间:
2020-04
期刊:
Proceedings. IEEE International Symposium on Biomedical Imaging
影响因子:
--
通讯作者:
Li G
Li G
中科院分区:
其他
文献类型:
--
作者:
Zhang X;Ding X;Wu Z;Xia J;Ni H;Xu X;Liao L;Wang L;Li G

文献摘要

参考文献

被引文献

相似文献

自闭症谱系障碍(ASD)是一种复杂的神经发育障碍,缺乏生物诊断标志物。因此,直接从脑成像数据中探索 ASD 识别一直是一个重要的课题。在这项工作中,我们提出了 Siamese 验证模型,使用 6 个月和 12 个月的皮质特征来识别 ASD。我们不是直接将测试对象分类为自闭症谱系障碍,而是确定其与已成功诊断的参考对象是否具有相同或不同的标签。然后,根据与所有参考对象的比较,我们可以预测测试对象的标签。将分类问题建模为验证框架的优点在于,它可以大大扩大训练数据规模,使我们能够以端到端的方式训练出更准确、更可靠的模型。此外,为了进一步提高分类性能,我们引入了路径签名(PS)特征,它可以捕获大脑发育的动态纵向信息,用于ASD识别。实验表明,与最先进的方法相比,我们提出的方法达到了最好的结果,即 87% 的准确度、83% 的灵敏度和 90% 的特异性。
Autism spectrum disorder (ASD) is a complex neurodevelopmental disability, which is lack of biologic diagnostic markers. Therefore, exploring the ASD Identification directly from brain imaging data has been an important topic. In this work, we propose the Siamese verification model to identify ASD using 6 and 12 months cortical features. Rather than directly classifying a testing subject is ASD or not, we determine whether it has the same or different label with the reference subject who has been successfully diagnosed. Then, based on the comparison to all the reference subjects, we can predict the label of the testing subject. The advantage of modeling the classification problem as a verification framework is that it can greatly enlarge the training data size and enable us to train a more accurate and reliable model in an end-to-end manner. In addition, to further improve the classification performance, we introduce the path signature (PS) features, which can capture the dynamic longitudinal information of the brain development for the ASD Identification. Experiments showed that our proposed method reaches the best result, i.e., 87% accuracy, 83% sensitivity and 90% specificity comparing to the state-of-the-art methods.
DOI: 10.4171/rmi/240
发表时间: 1998-01-01
影响因子: 1.2
作者:
Lyons, TJ
通讯作者: Lyons, TJ
DOI: 10.1002/hbm.24636
发表时间: 2019-09-01
影响因子: 4.8
作者:
Wu, Zhengwang;Wang, Li;Shen, Dinggang
通讯作者: Shen, Dinggang
使用全卷积循环网络学习空间语义上下文,用于在线手写中文文本识别
DOI: 10.1109/tpami.2017.2732978
发表时间: 2018-08-01
影响因子: 23.6
作者:
Xie, Zecheng;Sun, Zenghui;Lyons, Terry
通讯作者: Lyons, Terry
DOI: 10.1007/978-3-030-00919-9_35
发表时间: 2018-09
期刊: Machine learning in medical imaging. MLMI (Workshop)
影响因子: --
作者:
Li G;Liu M;Sun Q;Shen D;Wang L
通讯作者: Wang L
DOI: 10.1016/j.nicl.2017.08.017
发表时间: 2018-01-01
影响因子: 4.2
作者:
Heinsfeld, Anibal Solon;Franco, Alexandre Rosa;Meneguzzi, Felipe
通讯作者: Meneguzzi, Felipe